No natural scale of colours

On p. 92 of Envisaging Information, ET points out that “the mind’s eye does not readily give an order to ROYGBIV”, and points out that shades of intensity are much easier to interpret — more or less obvious once pointed out, but not at all obvious to people who haven’t thought about it.

Now that it has become so easy to generate coloured pictures with computers, this sensible advice is more and more widely ignored. At the moment I am assessing a doctoral thesis in which the author has a graph with 1600 spots colour-coded for degrees of heterosis from 0 to more than 200%. (Heterosis is a genetic measure of the extent to which a child deviates from the average for the two parents). It’s not a bad graph from the point of view of information density, but unfortunately the two darkest colours (out of five) refer to “0” and “more than 200%”, with three fainter colours used for the intermediate degrees (one of them so faint as to be almost invisible). As a result it is very difficult for the eye to pick out much in the way of a pattern. With shades of grey it could have been so much better, and the author could have distinguished between more than five levels in a way that would have been intuitively clear.